THERMAL DECOMPOSITION and EXPLOSION of AMMONIUM PERCHLORATE and AMMONIUM PERCHLORATE PROPELLANT up to 50 KILOBARS (5.0X10' N/M2)

Home , Ammonium perchlorate, Chemical decomposition, Thermal decomposition

NASA TECHNICAL NOTE NASA TN D-6013

cr) c- 0 liom COPY -? n Am KIR!I"D I I- 4 m 4 z

THERMAL DECOMPOSITION AND EXPLOSION OF AMMONIUM PERCHLORATE AND AMMONIUM PERCHLORATE PROPELLANT UP TO 50 KILOBARS (5.0X10' N/m2)

by Huns R. Voelkl Lewis Research Center Cleueland, Ohio 44135

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. C. SEPTEMBER 1970 TECH LIBRARY KAFB, NM

- ... __ - 0132bb4 1. Report No. 2. Government Accession No. 3. Recipient's Catalog IVO. NASA TN D-6013 - -.. 4. Title and Subtitle THERMAL DECOMPOSITION AND EXPLOSION OF 5. Report Date September 1970 AMMONIUM PERCHLORATE AND AMMONIUM PERCHLORATE 6. Performing Organization Code PROPELLANT UP TO 50 KILOBARS (5. OXlO9 N/m2)

7. Author(s) 8. Performing Organization Report No. Hans R. Voelkl E-4568 10. Work Unit No. 9. Performing Organization Name- and Address 129-03 Lewis Research Center 11. Contract or Grant No. National Aeronautics and Space Administration

Cleveland, Ohio 44135 13. Type of Report and Period Covered I12. Sponsoring Agency Name and Address Technical Note - National Aeronautics and Space Administration 14. Sponsoring Agency Code Washington, D. C. 20546 -. 15. Supplementary Notes

L - - . . - -. 16. Abstract The rates of thermal decomposition of ammonium perchlorate and an ammonium perchlorate 9 9 9 2 solid propellant at 15, 25, and 50 kilobars (1.5X10 , 2.5X10 , and 5.0XlO N/m ) pressure were studied in a cubic anvil press. The data were correlated by first order rate equations to obtain the temperature variation of the specific reaction rates and the apparent activation energies. Explosion limits of pure ammonium perchlorate are included to 30 kilobars (3. OXlO9 N/m2). Electrical resistance measurements of this salt at high pressure were also made.

.- .~ 18. Distribution Statement Thermal decomposition Activation energy Unclassified - unlimited Ammonium perchlorate Explosion limits Kilobar Rate equation ~- -_ 19. Security Classif. (of this report) 20. Security Classif. (of this page) Unclassified Unclassified $3.00 -~

'For sale by the Clearinghouse for Federal Scientific and Technical Information Springfield, Virginia 22151 THERMAL DECOMPOSITION AND EXPLOSION OF AMMONIUM PERCHLORATE AND AMMONIUM PERCHLORATE PROPELLANT UP TO 50 KILOBARS (5. Ox109 N/m2) by Hans R. Voelkl Lewis Research Center

SUMMARY

Thermal decomposition of ammonium perchlorate (AP) and an AP based propellant was studied at high pressure in a cubic anvil press. Preexplosion kinetics of the thermal decomposition of AP and propellant were determined between 266' and 337' C at 15, 25, and 50 kilobars (1. 5x109, 2. 5x109, and 5.0XlO 9 N/m 2 ). First-order kinetics were observed in all cases. Arrhenius plots of reaction-rate constants at different decomposition temperatures allowed the calculation of activation energies. For AI? the energies 5 9 2 ranged from 40.4 kilocalories per mole (1.69~10J/mole) at 15 kilobars (1.5~10N/m ) to 57.2 kilocalories per mole (2. 39x105 J/mole) at 50 kilobars (5.0~109 N/m 2 ). Activa- tion energies for propellant were calculated to be 30.3 kilocalories per mole (1.27X10 5 J/mole) at 15 kilobars (1.5~10' N/m2) and 48.5 kilocalories per mole (2.03~10~J/mole) at 50 kilobars (5.0X109 N/m2). Decompositions of AP were explosive at a heating rate of 10' C per minute. Explo- sion limit curves (T against P) showed explosion temperatures increasing with increas- 9 2 ing pressure up to 30 kilobars (3.0xlO N/m ). Addition of small amounts of aluminum and copper chromite powders to AP and propellant reduced explosion temperatures from 10' to 60' C depending upon the pressure. Electrical resistance of AP pellets at high pressure with temperature variation showed no discontinuities. This suggested that no phase transformations occurred.

INTRODUCTION

There is a great technological interest in ammonium perchlorate (AP)because of its extensive use in solid propellant rockets. In this work AP is subjected to pressures and temperatures where many materials undergo phase transformations. Perhaps the best known case of a phase transformation is the formation of diamond from graphite (ref. 1). A phase change in AI? would give a material with drastically different proper- ties, which, in turn, might change the flame characteristics and burning rate. The thermal decomposition of AP has been studied in great detail at atmospheric and reduced pressure. Studies of this material show several unique features. Bircumshaw and Newman (ref. 2) found that below 300' C the decomposition does not go to completion, but stops after about 30 percent of the salt has decomposed, leaving a residue with the same chemical composition. The kinetics and mechanism of decomposition are temperature dependent (ref. 3). Galwey and Jacobs (ref. 4) suggested that below 300' C an electron transfer mechanism involving radical formation is responsible for the decomposition. Galwey and Jacobs (ref. 5) also found that temperatures above 350' C cause the decomposition to go to completion. At these higher temperatures the reaction can be explained by either a proton transfer mechanism or rupture of chlorine- oxygen bonds. Above 440' C the decomposition is explosive (ref. 6). Deflagration characteristics of pure AP have been studied by means of a closed-bomb strand burning tech- 6 nique at pressures from 1000 to 23 000 psi (6.895XlO to 1. 586X108 N/m2) (ref. 7). This investigation deals with the influence of pressure on the thermal decomposition of AP and an AP propellant in a multianvil high pressure apparatus. Preexplosion kinetics of AP and propellant are determined between 266' and 337' C at 15, 25, and 9 9 2 50 kilobars (1. 5x1O9, 2.5xlO , and 5.0xlO N/m ). Specific reaction-rate constants are calculated from first-order equations. These data permit calculation of the energy of activation for the decompositions. Explosion limits of AP are determined by heating compressed pellets of the salt at a slow, constant rate at fixed pressure. Data are given for the effect of finely divided aluminum powder and copper chromite additives on explosion limits. In addition, some data on the electrical resistance of AP at high pressure were obtained.

EXPERIMENTAL MATERIALS AND PROCEDURE

Materia I s

Ordnance grade AP (99.5 percent minimum purity), twice recrystallized from water, finely ground was used in this study. Bismuth, barium, and thallium, in wire form for calibration experiments, were high purity elements (>99.999 percent). Pyrophyllite (A123 0 .4Si02. H20) is a naturally occurring substance and was used as received. The propellant used in this work contained 80 percent by weight polybutadiene-acrylic acid and 20 percent by weight AP. The AP was a mixture of 20 percent fine material (ll-ym average particle diameter) and 80 percent coarse (88-y m average particle diameter).

2 Apparatus

All high pressure experiments performed in this study were carried out with a multianvil pressure apparatus of cubic configuration. The apparatus consists essentially of six hydraulically operated pistons to which are attached square-faced tungsten carbide anvils that form a cubic volume. Each piston is pushed inward along its axis against the material to be pressurized. The sample to be compressed is put into a pyrophyllite cube which is 10 to 15 percent larger in edge dimension than the 0.89-inch (2.26-cm) square anvil face. When the anvils touch the pyrophyllite cube, there are gaps between the anvils which allow forward movement of the anvils to compress the sample. As the anvils advance, some of the pyrophyllite extrudes into the gaps to form the gasket structure which gives support to the anvils and also absorbs some of the thrust load. Details of the apparatus have already been described (ref. 8).

P ressu re Ca I ib ratio n

A pressure calibration of the apparatus must be determined before any pressure runs are made. Calibration is necessary because friction and pressure in the gasket areas absorb an unknown amount of the thrust force on the anvil faces. Therefore, the overall force-area ratio is not the true pressure in the center of the pressure chamber. The actual pressure calibration is accomplished by placing a material in the pressure chamber which is known to undergo an abrupt shift in electrical resistance or in volume at a definite pressure and determine the force at which the "known" transitions occur. The known transition pressures have been determined previously in a free-piston apparatus in which the pressure can be calculated accurately from the force-area ratio. The pressure calibration of the apparatus in this work was determined by measuring the load required to convert BiI to BiII, BiII to BiIII, TlII to TlIII, BaII to BaIII, and BiV to BiVI. The bismuth, thallium, and barium were in wire form enclosed in cast silver chloride as shown in figure 1. The silver chloride acts as a nearly hydrostatic pressure transmitting medium. The accepted values of transition pressures for bismuth, thallium, and barium are those of Kennedy and La Mori (ref. 9). A typical transition curve of bismuth is given in figure 2 which shows sharp changes at the transition point when relative resistance is plotted against load. The electrical resistance of the calibration wire is followed by observing the potential drop across two opposed anvils at constant current. The resistance due to the anvils and copper leads is negligible. At 9 6 the point where BiI BiII, the pressure required is 25 410k95 bars (2.541XlO zt9.5X10 2 - N/m ). Figure 3 shows a calibration curve giving the transition values for bismuth, thallium, and barium plotted against press load. Load-resistance characteristics for

3 rCalibration wire

power supply

decade box

Figure 1. - Schematic diagram for pressure calibration.

rBiI- BiII 4 ,/’

TBiII - BiIII 2- r BiV - BiVI

I I I 1 I I 0 2000 4000 6000 8wO 10000 12000 14 000 Ram pressure, psi

IL-’ 1 5 10x107 Ram pressure, N/m2 Figure 2. - Bismuth pressure transition curve.

loor BiV - BiVI 80-

60 -

I l--UU1 0 2000 4000 6000 8000 10 OOO 12 000 14 000 Ram pressure, psi

I 10?107 1 5 Ram pressure, N/m2 Figure 3. - Pressure calibration curve.

4 the calibration wires were observed on an X-Y recorder, the load signal being derived directly from a calibrated standard pressure cell and resistance from the potential drop across the wire at constant current.

P reexplosio n K i netics

AP samples for the kinetic study were prepared in the form of 3/8-inch- (0.95-cm-) diameter disks. They were made by compaction of 140 milligrams of polycrystalline powder in a pellet mold at 180 000 psi (1.241~109 N/m 2 ) at room temperature. An AP pellet was then placed between tantalum foil backed up with stainless steel disks in a boron nitride container surrounded by a graphite cylinder as shown in figure 4. The

2.25 in. I_ (5.715 cm)

Boron nitride-, -- - _-_- Graphite heater Stainless steel------_-_-Thermocouple well

------Copper tab /--- Tantalum foil -- Insulator - -

0.625 in. (1.587 cm) , Silver foil'

Figure 4. -Cross sec on of high pressure cell assenibly.

graphite was resistively heated at the rate of 10' C per minute by means of a welding transformer driven by a variable speed motor. Current was passed through two opposed anvils and led to the graphite heater by means of copper tabs. Massive pyrophyllite was used for specimen containers and as a solid pressure transmitting medium. Integral preformed gaskets were machined on the pyrophyllite cubes to minimize sample deformation during initial stages of pressure application (ref. 10). Sample temperatures as determined by thermocouple measurements at high pressure were kept constant within 0.5' C. Sample temperatures were continuously monitored on an X-Y recorder. The experiment was performed by compressing an AP pellet to the desired pressure,

5 raising the temperature to the desired level, and holding these conditions for a pre- scribed time. Then temperature and pressure were reduced to ambient conditions and the sample weighed to determine weight loss. The resulting weight loss records were analyzed to determine the dependence of decomposed AP upon time.

Explosion Limits

In these experiments the weight of AP was reproduced in different runs to 140A milligrams. Samples were confined between 3-mil-thick tantalum foil. The AP pellets were brought to the desired pressure at room temperature and then heated at the rate of 10' C per minute. The temperature of explosion at the various pressures was determined by recording the temperature at which a detonation was clearly heard during the heating of the sample. As a result of the explosions, pyrophyllite was ejected from the gasket areas of the container with considerable force. Some explosions, particularly those at the higher pressures were of sufficient violence to damage or destroy the carbide anvils. For this reason explosion limits were determined only to 30 kilobars (3. OXlO9 N/m2).

Elect r ica I Resi sta nce

Resistance measurements were made with an electrometer equipped with power supply and shunt. Contact with the AP pellet was made with tantalum foil backed up with stainless steel disks. The circuit to the exterior of the pyrophyllite cube was completed with copper wire and silver foil. Resistance was measured through two opposed anvils. The resistance of the electrical circuit excluding the AP pellet was negligible. Temper- ature was measured with Chromel/Alumel thermocouples placed near the perchlorate sample as shown in figure 4. The output of the electrometer and thermocouple were fed into X-Y recorders to plot the changes in resistance with temperature.

RESULTS AND DISCUSSION

P reexplosion Kinetics (Ammonium Perch lorate)

Rate data in the temperature range 282' to 337' C were obtained for the thermal 9 decomposition of AP at 15, 25, and 50 kilobars (1. 5x109, 2.5~10'~and 5.0XlO N/m2). A first-order reaction for the disappearance of the salt was assumed since plots of the logarithm of the mole fraction of remaining AP against time produced approximately 6 TABLE I. - DATA FOR PRESSURE RUNS (AMMONIUM PERCHLORATE) 2 Pressure, kbar (N/m )

15 (1.5~10~) 25 (2.5~10') 50 (5.0~10~)

~ smper- Time, Mole Temper- Time, Mole Temper- Time, Mole ature, min fraction ature, min fraction ature, min fraction OC remaining OC emaining OC -emaining

~ 2 82 6 0.920 293 5 0.925 3 12 5 0.923 11 .a65 10 .856 20 .728 16 .802 25 .697 40 ,544 31 .652 40 .541 60 .397 51 .520 60 .405 a5 .278 ai .370 80 .309 ~ 324 4 0. a35 295 1.5 0.958 305 5 0.858 14 .548 6.5 .815 10 .722 29 .306 16.5 .615 20 .504 44 . 158 31.5 .443 35 .319 59 .085 41.5 .353 60 . 136

~ 337 5 0.554 61.5 .212 3 12 1 0.943 10 .300 305 1 0.922 6 .726 15 .178 6 .653 16 .425 20 .096 16 .399 26 .243 26 .265 41 .113 - 46 .099 3 24 4 0.597 316 2.5 0.752 9 .319 7.5 .451 14 .182 12.5 .224 19 .091 17.5 .092

~ straight lines. The data for these pressure runs are tabulated in table I. Figures 5 to 7 show the results obtained at 15, 25, and 50 kilobars (1.5X10 9 , 2. 5x109, and 5.0xlO 9 2 N/m ). The decompositions in most cases were carried out to at least 80 percent completion. Values of the specific reaction-rate constants k were calculated from the slopes of the aforementioned plots and are shown in table 11. From reaction-rate theory (ref. ll), the rate constant k may be written as

-E/RT k = Ae where A is a constant independent of the nature of the reaction, E is the activation energy, R is the universal gas constant, and T is absolute temperature. Figure 8 shows plots of log rate constants against 1/T for a number of runs at

I I I I I I 111

-.40

-. 80

-1.20

0 -S -1.60

-2.00

-2.40

-2. 80

a 6

Time, miri Figure 5. - Fraction of undecomposed ammonium perchlorate (AP) at 15 kilobars (1. 5x109 N/m2).

I -2. 80 I k4

a 6

Figure 6. - Fraction of undecomposed ammonium perchlorate (AP) at 25 kilobars (2. 5x109 Nlm2).

9 Temperature,

0 -c

-2.40-

0 10 20 30 40 50 60 70 80 90 Time, min

Figure 7. - Fraction of undecomposed ammonium perchlorate (API at 50 kilobars (5.0~10~N/mZl.

10 TABLE 11. - SPECIFIC REACTION RATES FOR DECOMPOSITION

OF AMMONIUM PERCHLORATE I

15 (1.5~10')

Temper- kverage reaction- Temper- Average reaction- Temper- Average reaction- ature, rate constant, ature, rate constant, ature, rate constant, c OC k, OC k, OC k, -1 sec-l sec-l s ec

282 2.12x10-~ 293 2.46~10-~ 3 12 2.54~10-~ 295 4.53~10-~ 305 5.50~10-~ 3 24 7.01~10-~ 305 1. OOX~O-~ 312 8.80~10-~ 337 1.92~10-~ 2.10~10-~ 3 16 1.77~10-~ 3 24 II

-Ammonium perchlorate -Propellant

-2.90

Y --0 -3.10

-3.30

-3.50 I

-3.70 iII 1.62 1.64 Reciprocal decomposition temperature, UT, K

Figure 8. - Arrhenius plots of thermal decompositions at high pressures.

I different temperatures and pressures which followed first-order kinetics. The data are represented by straight lines for the three pressures from which activation energies 5 were calculated. The values found were 40.4 kilocalories per mole (1.69~10 J/mole) 9 at 15 kilobars (1.5~10 N/m2), 46.3 kilocalories per mole (1.94~10~J/moIe) at 25 kilobars (2.5~109 N/m 2 ), and 57.2 kilocalories per mole (2.39XlO5 J/mole) at 50 kilobars (5.0~10~~/m~). The rate equation for the thermal decomposition of AI? at 15 kilobars (1.5~109 N/m2) was calculated to be c

k = 2.05X10 12 ,-40 400/RT sec-l

The rate equation at 25 kilobars (2.5X10 9 N/m 2 ) was found to be

k = 2.19X10 14 ,-46 300/RT sec-l

9 2 At 50 kilobars (5.0XlO N/m ) the rate equation was calculated to be

k = 7.42X10 17 ,-57 200/RT sec-l

P reexplosion Kinetics (A P P rope1lant)

Kinetic data in the temperature range 266’ to 308’ C were obtained for the thermal decomposition of an AP solid propellant at 15, 25, and 50 kilobars (1.5X109 , 2.5~109 , 9 2 and 5.0xlO N/m ). Results similar to those for pure AP decomposition were obtained. Plots of the logarithm of the fraction of undecomposed propellant against time produced approximately straight lines. The data are tabulated in table III. Specific reaction-rate constants were calculated from the slopes of the aforementioned lines and are shown in table IV. Arrhenius plots of log rate constants against 1/T at 15, 25, and 50 kilobars 9 9 9 2 (1.5~10, 2.5xlO , and 5.0XlO N/m ) are shown in figure 8. Straight lines were obtained which allowed the calculation of activation energies of decomposition. The rate equation for the thermal decomposition of the propellant at 15 kilobars (1.5~109 N/m 2 ) was calculated to be

k= 1.70~109 e -30 300/RT sec-l

9 2 At 25 kilobars (2.5~10 N/m ) the rate equation was found to be

11 ,-xi ~OO/RTsec-i k = 4.66X10 12 TABLE 111. - DATA FOR PRESSURE RUNS (PROPELLANT) 2 Pressure, kbar (N/m )

15 (1.5~10~) 25 (2.5~10’)

Temper- rime, Mole remper- Time, Mole remper- Time, Mole ature, min fraction ature, min fraction ature, min fraction OC :e maining OC remaining OC remaining

266 5.5 0.766 277 5 0.748 2 89 5 0.752 15.5 .477 10 .589 10 .571 30.5 .193 15 .449 15 .426 45.5 .089 25 .256 20 .300 25 .233 2 78 1 0.910 2 89 5 0.558 6 .615 10 .332 297 5 0.613 11 .407 15 ,189 10 .353 16 .270 20 . €06 15 .232 23.5 .145 20 .123

~~ 301 3 0.497 287 1 0.860 5 .340 308 2.5 0.550 6 .447 7.5 .178 5 .287 11 .205 10 . 106 7.5 ,133 16 .080 io .081

TABLE IV. - SPECIFIC REACTION RATES FOR DECOMPOSITION OF PROPELLANT Pressure, kbar (N/m 2 ) 1 15 (1.5~10~N/m2) 25 (2.5~10’) 50 (5.0~10~) Temper- Average reaction- Temper- I Average reaction- ature, rate constant, ature, rate constant, ature, rate constant, OC k, OC k, OC sec-1 sec-l

266 7.96~10-~ 277 289 2 78 1.36~10-~ 2 89 297 287 2.31~10-~ 301 308

13 9 2 At 50 kilobars (5.0xlO N/m ) the rate equation was calculated to be

15 ,-48 500/RT sec-l k = 8.05X10

Figure 8 shows that the effect of increasing pressure is to reduce the rate of thermal decomposition. For an increase in applied pressure it is necessary to use a higher temperature in order to obtain a decomposition rate comparable to one at a lower temperature. A decrease in reaction rate with increasing pressure may be expected on the 8 basis of Le Chatelier's principle. Since the decomposition of AP yields gaseous prod-

ucts occupying a greater volume than the starting material, the application of external d pressure would favor retention of reactant, that is, retard the decomposition. Other workers (refs. 12 and 13) have studied the effect of high pressure on the thermal decomposition of explosives. Increasing pressure altered the rate of decomposition; in most cases a decrease in rate was observed. Bridgman (ref. 14) studied the effect of high mechanical stress up to 100 000 atmospheres on solid explosives. He found that a combination of high temperature with high stress was needed to produce an 9 2 explosion. In this work, pressures up to 80 kilobars (8.0X10 N/m ) failed to produce explosions or any decomposition in AP at room temperature. The variation of activation energy for the decomposition of AP and AP propellant at 9 9 2 15, 25, and 50 kilobars (1.5X10 , 2. 5x1o9, and 5.0XlO N/m ) is shown in figure 9. The energy barrier for decomposition increases with increasing pressure. Extrapola- tion of the AP curve back to atmospheric pressure gives an activation energy of 29 to 30 kilocalories per mole (1.21XlO5 to 1.255XlO 5 J/mole). This value is in very good agreement with the results obtained by Galwey and Jacobs (ref. 3) for the decomposition of AI? pellets at 1 atmosphere (28 to 31 kcal/mole or 1.17 to 1.30 J/mole). Similar values were obtained by Bircumshaw and Newman (ref. 2). It should be pointed out that the activation energies calculated at atmospheric pressure were derived from different kinetic equations. Those workers used the Prout-Tompkins (ref. 15) and Avrami- Erofe'ev equations (refs. 16 and 17). The data for the decompositions of the present study were fitted to the Arrhenius equation. The data did not fit either the Prout- Tompkins or the Avrami-Erofe'ev equations.

Explosion Limits

Explosion limits for pure AP and AP containing aluminum powder and copper chro- 9 2 mite up to 30 kilobars (3.0XlO N/m ) pressure are shown in figure 10. The areas above the curves represent explosive regions. In all cases increasing pressure results in higher explosion temperatures. Addition of finely divided aluminum and copper chro-

14 2.5 !? 60

2 2.0 - ....E 7 w' 45 -W s 0 P - W Wc mU *40 c0 w- .-w m> s ._ P -U W 1.5 -c : 35 ._0 c m ._ cu a 30

25 1 ; I I I 20

Pressure, kbar

Figure 9. -Variation of activation energy of decomposition with pressure.

15 0 Pure ammonium perchlorate o AP + 0.25 wt. % copper chromite o AP + 1.0 wt. % aluminum A Composite propellant 370

360

350

340

- 330

a;L 2 320 L CLa,

cE c 310 ._0 -0“l a x 300

290

280

270

260 Pressure, kbar

I.--. L - ~ 1 ~~ - -1 0 1 2 3x109 Pressure, N/m2

Figure 10. -Explosion limits of ammonium perchlorate (API and propellant at high pressure.

mite causes a considerable reduction in explosion temperature. I Since the decomposition of AP is an exothermic reaction, temperature is the main factor which distinguishes the conditions of explosion from those of slow decomposition. i An explosion results when the rate of heat production by chemical decomposition exceeds the rate of heat loss to the surroundings. It is clear from the results that the effect of pressure is to hinder the initiation of explosion. This fact could be predicted on the basis of Le Chatelier’s principle. Since an explosion is a chemical reaction yielding products at high pressure, the external application of pressure should hinder explosion.

16 Since aluminum and copper chromite act as catalysts for the decomposition of ammonium perchlorate, it should be expected that lower initiation temperatures would be required. The reason for the higher explosion temperatures of the propellant as com- pared to pure AP may be due to the binder (80 percent by weight of the propellant) acting as an inhibitor.

. Elect r ica I Resistance The temperature dependence of resistance of AP at high pressure is given in figure 11. Isobar I shows the resistance behavior for the first heating at 10 kilobars (1.0~109 N/m 2 ); isobar I1 represents the cooling curve. Additional temperature cycling

io (1.0~1091

104 I I I I I 25 75 125 175 225 275 325 375 Temperature, "C

Figure 11. - Electrical resistance of ammonium perchlorate at high pressure.

I 9 2 at 10 kilobars (l.Oxl0 N/m ) is also represented by isobar II. Isobars I11 and IV show resistance at 20 and 30 kilobars (2.0XlO 9 and 3.0XlO 9 N/m 2 ), respectively. The type of resistance minimum in isobar I at approximately 125' C is usually associated with a phase transformation. However, this probability was ruled out because the cooling curve and temperature cycling showed no minimum. Also, X-ray powder diagrams of recovered AP were identical with starting material. The density of AP samples prepared in the pellet mold was 1.92 grams per cubic centimeter. After com- pression to 10 kilobars (l.OXl0 9 N/m 2 ) and heating to 300' C, the density increased to the single crystal value of 1.95 grams per cubic centimeter. Thus, the resistance minimum is attributed to voids in the sample which are removed during the first com- r pression and heating.

24v 104 1 I. J L6 2. 0 2. 4 2. e 3.2 3.6~10-~ Temperature, lIT, K

Figure 12. - Electrical resistance of ammonium perchlorate as function of temperature.

18 Tantalum foil was the best material for electrical contacts in resistance measurements. This material was found to be inert to AP under the high temperatures and pressures used in this study. Platinum and molybdenum were unsuitable for making resistance measurements. Under the aforementioned conditions these metals were attacked by AP and gave unreproducible resistance curves. Ammonium perchlorate undergoes a phase transformation in the temperature range 240' to 2'70' C at atmospheric pressure (refs. 18 and 19). The crystal change is from the orthorhombic to the cubic form. Maycock, Verneker, and Gorzynski (ref. 20) studied the electrical conductivity of AP in the temperature range 60' to 280' C. Their plot of conductivity as a function of temperature showed a "knee" at 255' C which is attributed to the change in crystal structure. The plot of electrical resistance against reciprocal temperature at 10, 20, and 30 kilobars (l.OXl09 , 2.0~10 9 , and 3.0~10 9 N/m 2 ) in figure 12 shows none of the discontinuities usually associated with phase changes. It is possible the application of high pressure prevents the transformation, since there is a 9.7-percent increase in volume associated with the transition. The temperature shift of the transition as a function of pressure can be calculated by the Clap eyron- C lausius equation (ref . 2 1):

Ldp = AH dT T(Vc - Vo)

where AH is the enthalpy of transition, T is the absolute temperature, Vc is the molar volume of cubic AP, equal to 5. 682X10-4 liter per gram (0.5682 cm3 /g), and Vo is the molar volume of orthorhombic AP, equal to 5. 128X10-4 liter per gram (0.5128 cm3 /g). Using 2.3 kilocalories per mole (9.62~103 J/mole) for the enthalpy of transition (endothermic) as measured by Markowitz and Boryta (ref. 22) and the factor for converting calories to liter-atmospheres

. 1.9 87 (cal)(deg-')( mole- ') or

I 82.05 (cm3-atm)(deg-l)(mole-') 3 1 = 9.8694 (cm -atm)(J- ) 8.314~10~(J)(deg- ')(mole- ') l one obtains

19 [2300 (cal)(mole-')] [O. 04129 (liter-atm)(cal-l)].. -dP = dT (513 deg) [5. 54xW5 (liter)(g-l)] [117.5 (g)(mole-'j] or

[9.62x106 (J)(mole-')]-~ ~ . [9.8694.- -(cm3-atm)(J-l)] -. - dp= dT (513 deg) [5. 54X10-2 (cm3)(g-l)] [117.5 (g)(mole-1 )J 1 .

= 28.4 (atm)(deg-l)

The reciprocal, dT/dp = 0.035 degree per atmosphere, indicates that an increase in pressure of d atmosphere raises the transition point 0.035 degree. A pressure of 10 kilobars (1.0~109 N/m 2 ) should raise the transition point approximately 350 degrees to about 590' C. This is obviously impossible, as shown by the explosion limit data at high pressure. High pressure X-ray measurements would be useful for determining the transition point of AP as a function of pressure. Such a method would clearly detect the orthorhombic to cubic transition.

SUMMARY OF RESULTS

The more significant results obtained from this study of solid ammonium perchlorate and an ammonium perchlorate based solid propellant at high pressures were as follows : 1. Thermal decompositions of AP and propellant at high pressure and elevated temperatures follow first-order kinetics and go to completion. Rate equations for the de- 9 composition of perchlorate between 282' and 337' C at 15, 25, and 50 kilobars (1.5~10, 9 9 2 2.5X10 , and 5. OXlO N/m ) were calculated. Rate equations for decomposition of propellant between 266' and 308' C at 15, 25, and 50 kilobars (1. 5x109, 2. 5x109, and 5.0X109 N/m2) were calculated. 2. Explosion limits for ammonium perchlorate were determined to 30 kilobars (3.0X109 N/m2). Increasing pressure raises explosion temperatures. Metallic additives lower explosion temperatures. 3. A composite propellant containing ammonium perchlorate has higher explosion temperatures than pure AP alone.

20 4. Electrical resistance measurements of compacted polycrystalline ammonium 9 perchlorate powder suggested no phase transformations up to 30 kilobars (3.0XlO N/m2) and 340' C.

Lewis Research Center, National Aeronautics and Space Administration, Cleveland, Ohio, July 20, 1970, 129-03.

REFERENCES

1. Bovenkerk, H. P., et al. : Preparation of Diamond. Nature, vol. 184, no. 4693, Oct. 10, 1959, pp. 1094-1098. 2. Bircumshaw, L. L. ; and Newman, B. H. : The Thermal Decomposition of Ammonium Perchlorate. I. Introduction, Experimental, Analysis of Gaseous Products, and Thermal Decomposition Experiments. Proc. Roy. SOC. (London), Ser. A, vol. 227, no. 1168, Dec. 21, 1954, pp. 115-132. 3. Bircumshaw, L. L. ; and Newman, B. H. : The Thermal Decomposition of Ammo- nium Perchlorate. 11. The Kinetics of the Decomposition, the Effect of Particle Size, and Discussion of Results. Proc. Roy. SOC. (London), Ser. A, vol. 227, no. 1169, Jan. 7, 1955, pp. 228-241. 4. Galwey, A. K. ; and Jacobs, P. W. M. : The Thermal Decomposition of Ammonium Perchlorate at Low Temperatures. Proc. Roy. SOC. (London), Ser. A, vol. 254, no. 1279, Mar. 8, 1960, pp. 455-469. 5. Galwey, A. K. ; and Jacobs, P. W. M. : High-Temperature Thermal Decomposition of Ammonium Perchlorate. J. Chem. SOC., pt. 1, 1959, pp. 837-844. 6. Galwey, A. K. ; and Jacobs, P. W. M. : The Thermal Explosion of Ammonium Perchlorate. J. Chem. SOC., pt. 4, 1960, pp. 5031-5033. b 7. Irwin, 0. R. ; Salzman, P. K. ; and Andersen, W. H. : Deflagration Characteristics of Ammonium Perchlorate at High Pressures. Ninth Symposium (International) on I Combustion. W. G. Berl, ed., Academic Press, 1963, pp. 358-365. 8. Zeitlin, Alexander: Equipment for Ultrahigh Pressures. Mech. Eng., vol. 83, no. 10, Oct. 1961, pp. 37-43. 9. Kennedy, G. C. ; and La Mori, P. N. : Some Fixed Points on the High Pressure Scale. Progress in Very High Pressure Research. F. P. Bundy, W. R. Hibbard, Jr., and H. M. Strong, eds., John Wiley & Sons, Inc., 1961, p. 304. 21 10. Samara, G. A. ; Henius, A. ; and Giardini, A. A. : A Study of Pressure Homogene- ity in the Hexahedral Anvil High-pressure Apparatus. J. Basic Eng., vol. 86, no. 4, Dec. 1964, pp. 729-735. 11. Glasstone, Samuel; Laidler, Keith J.; and Eyring, Henry: The Theory of Rate Processes. McGraw-Hill Book Co., Inc., 1941. 12. Bowden, F. P.; Evans, B. L.; Yoffe, A. D.; and Yuill, A. M.: The Influence of High Pressure on Thermal Explosion and the Decomposition and Detonation of Single Crystals. Disc. Faraday SOC., no. 22, 1956, pp. 182-187.

13. Ryabinin, Yu. N. : The Influence of Pressure on the Velocity of Thermal Decompo- + sition of Explosives. Doklady Akad Nauk SSSR, vol. 58, 1947, pp. 245-248. 14. Bridgman, P. W. : The Effect of High Mechanical Stress on Certain Solid Explo- sives. J. Chem. Phys., vol. 15, no. 5, May 1947, pp. 311-313. 15. Prout, E. G. ; and Tompkins, F. C. : The Thermal Decomposition of Silver Per- manganate. Trans. Faraday SOC., vol. 42, June-July 1946, pp. 468-472. 16. Avrami, Melvin: Kinetics of Phase Change. I. General Theory. J. Chem. Phys., vol. 7, no. 12, Dec. 1939, pp. 1103-1112. 17. Erofe'ev, B. V. : Generalized Equation of Chemical Kinetics and Its Application in Reactions Involving Solids. Compt. Rend. Acad. Sci. URSS, vol. 52, 1946, pp. 511-514. 18. Hueckel, Walter (L. H. Long, trans. ): Structural Chemistry of Inorganic Com- pounds. Vol. 2. Elsevier Publ. Co., 1951, pp. 667-670. 19. Vorlaender, D. ; and IQascht, Erich: New Forms of Perchlorates. Ber. deut. chem. Ges., vol. 56B, 1923, pp. 1157-1162. 20. Maycock, J. N.; Verneker, V. R. Pai; and Gorzynski, C. S., Jr.: Electrical Conductivity of Ammonium Perchlorate. Solid State Comm., vol. 5, no. 4, 1967, pp. 225-227. 21. Glasstone, Samuel: Textbook of Physical Chemistry. Second ed., D. Van Nostrand Co., Inc., 1946. 4 22, Markowitz, M. M. ; and Boryta, D. A. : Some Aspects of the Crystallographic Transition of Ammonium Perchlorate. ARS J. vol. 32, no. 12, Dec. 1962, pp. I 1941-1942.

22 NASA-Langley, 1970 - 21 E-4568